The objective of this program is to elucidate the mechanisms of nucleic acid cleavage that proceed through radical intermediates by unambiguously generating putative reactive intermediates within oligonucleotides. Reactive intermediates are generated photochemically from synthetically novel, modified nucleosides which are incorporated site specifically in the polymer. Reaction pathways will be investigated through a combination of product studies, isotopic labeling (radioactive and stable) and polyacrylamide gel electrophoresis. DNA strand scission is believed to be a major cause of cell deactivation by ionizing radiation (which produces hydroxyl radical), and a number of natural products that exhibit antitumor activity. Radical intermediates are also postulated in oxidative cleavage processes mediated by synthetic chemical agents (ie. Fe.EDTA and copper phenanthroline). Detailed mechanistic studies of these chemical processes are often hindered by the structural complexity and limited amounts of the nucleic acid substrate available, as well as the myriad of reaction pathways initiated by some cleavage agents. Independent generation of reactive intermediates offers the advantage of minimizing the number of reaction pathways that are active in any given experiment. This is an experimental approach often pursued by physical organic chemists, which to our knowledge has never been applied to the study of processes involving nucleic acids. Deconvolution of complex nucleic acid cleavage processes could provide the impetus for the design of new antitumor drugs. The proposal put forth within the radiation biology community that a single hydroxyl radical may produce a DNA double strand break via interchain hydrogen atom transfer is a striking example of how this experimental approach could provide direction for the design of antitumor drugs. Specific generation of individual radical intermediates will make possible the determination of their respective abilities to effect complementary strand cleavage, and provide target lesions for drug designers to aim for.